Monday, 2018-07-23

*** tpb has joined #photonsdi00:00
*** Bertl_oO is now known as Bertl_zZ04:08
*** se6astian|away is now known as se6astian07:09
*** RexOrCine|away is now known as RexOrCine10:02
se6astianfelix_: are XC7A35T or XC7A50T pin compatible?14:59
se6astianwhich one did you choose ?14:59
*** Bertl_zZ is now known as Bertl15:07
felix_yep, they are fully pin-compatible15:34
se6astiangreat, thanks15:34
se6astianfelix_: would you still need the old 34C3 state plugin module for development now?15:34
felix_i chose the 35t and i'd say that the 15t might be sufficient, but it's not worth cutting corners there, since it might cause problems in the future and the 35t doesn't cost that much15:34
se6astianthere is a customer interested in purchasing one15:34
felix_when i have the new module, i don't need the old one any more15:35
felix_i'll primarily support the new module though15:36
se6astianthe customer would like the old one because he needs it immediately - nothing is decided yet but I am evaluating possiblities15:36
felix_so it would be good that the customer will get the new version as replacement when the new version is ready15:36
se6astianflorret also has one IIRC15:36
se6astianyes that is the plan of the customer15:36
felix_there are 3; i have two, but one is not working15:37
felix_and the current implementation is very hack-ish and doesn't work too well without poking it a few times15:37
felix_well, current aka old15:37
se6astianI think the customer wants to develop a new implementation15:38
se6astianand needs hardware for it15:38
se6astianthere might be a good overlap with what we do so we might benefit from each other15:38
felix_if you want me to ship you the working module, you shoud tell me that until tomorrow late afternoon, since i'm away from wednesday to sunday15:38
se6astianbut I dont know much yet, was just a 10 minute phone call15:38
se6astiando you have any development with the old module planned still?15:39
felix_sadly photonsdi isn't currently in a releaseable shape; it's on my todo list, but i spent much more time on designing the hardware than i planned15:39
felix_the parts of photonsdi i mainly worked on aren't hardware-dependent15:40
felix_well, the new board needs a new pin mapping file, but apart from that the underlying hardware doesn't matter too much at the moment15:40
se6astianso you expect no bottleneck/problem/delay if you send us the old module now right?15:41
felix_no, i don't expect any bottleneck for me from that15:41
felix_most of the existing code still isn't in the try-on-hardware-stage15:41
se6astianright15:41
se6astianthen please indeed ship the module to us just in case15:42
felix_ok15:42
felix_i need the address where the board needs to be shipped to15:42
se6astianfiguring that out currently15:42
felix_i include the 3d-printed card holder and the hdbnc-bnc adapter15:43
se6astiangreat15:43
felix_should i test it again before shipping or will you do that? last time i used the setup it worked flawlessly15:43
felix_oh and should i also include the non-working board?15:44
se6astianthats what I just asked herbert15:44
felix_ok15:44
felix_i've added the the shipment task on my todo list for tomorrow with highest priority :)15:45
se6astianthanks, I will get back to you with more info as soon as I get it15:47
felix_on the 35t/50t question: i used more than the number of decoupling capacitors needed for the 35t, but i'm not sure if it are as many as suggested for the 50t. if the 50t isn't completely full, it should be no problem. but using the 50t is an overkill as the 15t might even be sufficient15:47
felix_ok15:47
Bertlfelix_: you have a non-working module?15:51
felix_yes15:52
Bertlwhat doesn't work/how did it get broken?15:52
felix_module c is the woring one and i have look which one was the broken one15:52
felix_uh, i think you tried to rework the broken one, but that didn;t help15:53
felix_b is the broken one15:53
Bertlokay, the 'problem' there is something with the MGTs?15:53
Bertlbecause I tested the module I reworked (and it worked fine on the LVDS side, which I could test)15:54
felix_well, i have b and c here and c is the one in the 3d printed holder and i didn't change boards after i used it the alst time15:54
felix_uh, good question; i don't remember what the exact problem was15:54
Bertlbut you did test it _after_ the rework?15:54
felix_at least i remember testing it and it didn't work15:55
felix_i probably wrote that here in the channel15:55
Bertlokay, thanks!15:55
felix_i currently don't have the test setup set up on my desk, so it would take me maybe half an hour to get to the point where i could test things again; if it was set up, i'd just try the card again. i still could do that, but it would be a bit inconvenient for me at the moment15:57
Bertlno problem, not necessary15:58
felix_kk15:59
*** se6astian is now known as se6astian|away16:40
*** se6astian|away is now known as se6astian17:19
se6astiancustomer is still undecided if he needs the hardware already17:23
se6astianand asks:17:23
se6astian"Do you have the Vivado project for the Chaos Congress demo somewhere?  That would be the most useful actually - it's show us if what we want to  do is feasible using that part or we have to go to a Zynq etc."17:23
se6astianfelix_:18:52
*** RexOrCine is now known as RexOrCine|away19:29
felix_se6astian: you should have access to the git where the setup from the congress resides in. it's not a vivado project though, but uses vivado in non-project mode21:27
felix_from wednesday to sunday i'm not in my office and from the 7th august on i'm two weeks in the us21:30
se6astianthanks21:32
se6astiandoesnt seem like its super urgent for the customer but I will report if I hear anything back21:33
se6astianin any case dont stress yourself in any way about this21:33
se6astianI told them you are leaving wednesday to sunday21:33
felix_just wanted to let you know about my schedule21:33
*** se6astian is now known as se6astian|away21:34
felix_wouldn't be a problem for me to ship you the boards tomorrow21:34

Generated by irclog2html.py 2.13.1 by Marius Gedminas - find it at mg.pov.lt!